You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
With the native file upload input, not all screen readers always announce the chosen file: JAWS and NVDA both with Chrome and Firefox, VO on macOS with Safari (Chrome seems to be fine).
Our solution currently doesn't announce the file name. But it probably should.
What it could say:
just the file name
the file name plus something like "selected"
the file name plus the page title (because the user had left the browser window just before, but it might say that anyway)
the file name plus the page title plus the input label (but it might say that anyway)
something else?
When using the native file input with a screen reader it usually reads quite a lot of things (I think because it got focus back to the browser window) which is quite verbose. It's not a good idea to emulate a specific screen reader directly because they all say wildly different things. It makes sense to say the gist of what they all say.
Why
To let screen reader users know they have successfully selected a file and which file it was.
Also to meet WCAG 2.2 SC 4.1.3 Status Messages.
Who needs to work on this
Accessibility specialist, maybe a content designer
Who needs to review this
We'll review the decision as a team in a decision log meeting
Done when
Options with their pros and cons documented
Researched what other implementations do
Decision made and logged
The text was updated successfully, but these errors were encountered:
selfthinker
changed the title
Decide what is announced when a file is chosen or not, including the file name [Anika to create issue]
Decide what to announce when a file is chosen
Jan 8, 2025
What
With the native file upload input, not all screen readers always announce the chosen file: JAWS and NVDA both with Chrome and Firefox, VO on macOS with Safari (Chrome seems to be fine).
Our solution currently doesn't announce the file name. But it probably should.
What it could say:
When using the native file input with a screen reader it usually reads quite a lot of things (I think because it got focus back to the browser window) which is quite verbose. It's not a good idea to emulate a specific screen reader directly because they all say wildly different things. It makes sense to say the gist of what they all say.
Why
To let screen reader users know they have successfully selected a file and which file it was.
Also to meet WCAG 2.2 SC 4.1.3 Status Messages.
Who needs to work on this
Accessibility specialist, maybe a content designer
Who needs to review this
We'll review the decision as a team in a decision log meeting
Done when
The text was updated successfully, but these errors were encountered: